Simulation and verification of discrete element parameter calibration of pulverized coal particles

• Published in: International journal of coal preparation and utilization Vol. 45; no. 6; pp. 1244 - 1263
• Main Authors: Zhang, Jinxia, Wang, Qiuyue, Niu, Fusheng, Yu, Xiaodong, Chang, Zhenjia, Wu, Fan, Zhang, Mengfei
• Format: Journal Article
• Language: English
• Published: Taylor & Francis 03.06.2025
• Subjects: angle of repose; discrete element method; parameter calibration; Powder particles
• ISSN: 1939-2699; 1939-2702
• DOI: 10.1080/19392699.2024.2379405

To enhance the simulation efficiency of the EDEM discrete element numerical simulation software for coal powder mineral particles, experimental data on the rest angle of coal powder particles were collected as the response variable. Five parameters related to coal powder particles within the Hertz-Mindlin no-slip contact model were selected for investigation. Initially, the Plackett-Burman experiment identified three parameters: collision restitution coefficient, static friction coefficient, and rolling friction coefficient, significantly influencing the resting angle. Subsequently, the steepest ascent experiment determined the significant parameter ranges as follows: the collision restitution coefficient ranged from 0.20 to 0.40, the particle-particle static friction coefficient ranged from 0.30 to 0.50, and the particle-particle dynamic friction coefficient ranged from 0.06 to 0.10. Finally, the second-order regression model for the angle of repose and the saliency parameters was developed and optimized using the Box-Behnken experimental design. The following parameter combination was obtained: the collision recovery coefficient was 0.30, the particle-particle static friction coefficient was 0.49, and the particle-particle dynamic friction coefficient was 0.09. This study aimed to compare the angle of repose between the simulation experiment and the actual experiment. The relative error was only 1.02%, indicating that the simulation experiment achieved the optimal parameter combination.